To accurately obtain the switching characteristics of insulated gate bipolar transistors (IGBT), this paper proposes an IGBT's transient analysis model (TAM) based on the finite state machine (FSM). Firstly, the IGBT switching process in the basic converter unit is segmented in the time domain, and the circuit equations among the variables are obtained based on Kirchhoff's law. Then, the switching mechanism and parameters of the IGBT are decoupled based on time segmentation, and clarifying state equations and the transition events for each stage. Subsequently, all the stages and events are modeled in an FSM, thus establishing the IGBT's TAM, and it was realized the Stateflow toolbox of Matlab/Simulink. Finally, some experiments were performed on IGBTs with different current and voltage. The results show that the TAM can accurately describe the switching characteristics of IGBT and is suitable for different voltage and current levels.